Candlestick Plotting with Python Libraries

Create interactive Candlestick charts of OHLCV data obtained from yfinance library using python libraries.

Introduction

In this post, we will be looking at creating candlestick charts, which are integral to traders’ daily routines, using Python libraries such as Mplfinance, Plotly and Bokeh, and look at customizing the plots, change themes and tweak some colors.
As this blog post is written on Quarto, it enables viewers to interact with the output plots. You’ll be able to pan and zoom with the charts created with plotly and bokeh!
So, now lets head on to fetch our data first.

Getting Data

We will make use of the yfinance library. Let’s fetch the Open-High-Low-Close and Volume data for Apple (AAPL) for the past 5 months.

Our data looks like this:

Code

import yfinance as yf
import pandas as pd
import warnings
warnings.simplefilter(action='ignore', category=FutureWarning)

data = yf.download("AAPL", period="5mo",auto_adjust=True)
data.head()

C:\Users\soumy\AppData\Roaming\Python\Python310\site-packages\yfinance\base.py:48: FutureWarning:

The default dtype for empty Series will be 'object' instead of 'float64' in a future version. Specify a dtype explicitly to silence this warning.

[*********************100%%**********************]  1 of 1 completed

	Open	High	Low	Close	Volume
Date
2023-08-28	179.853052	180.352394	178.315085	179.952927	43820700
2023-08-29	179.463564	184.656719	179.263830	183.877747	53003900
2023-08-30	184.696681	187.602856	184.496947	187.403107	60813900
2023-08-31	187.592859	188.871173	187.233332	187.622818	60794500
2023-09-01	189.240698	189.670125	188.032284	189.210739	45732600

Plots

Now we will head straight into plotting. We will start with mplfinance, a special module for finance charts of matplotlib.

mplfinance

Creating candlestick charts with mplfinance is a simple one line code:

Code

import mplfinance as mpf

mpf.plot(data, type="candle", volume=True, style="yahoo",
         figsize=(6,3),title="AAPL Candlestick with Mplfinance",
        ylabel="Price [USD]")

This plot uses the “yahoo” style. We can check the available style using the command mpf.available_styles(). Now let us look at a few other available styles below:

Code

mc_list = ["binance","blueskies","starsandstripes","brasil"]

fig = mpf.figure(figsize=(6,6))
ax_dict = {}

for i,st in enumerate(mc_list):
    ax_dict[i]=fig.add_subplot(2,2,i+1, style=st)
    mpf.plot(data[:50], type ="candle", ax=ax_dict[i], axtitle=(st),xrotation=0)

And now we will tweak and play with the candle colors and background. First we will set the base mplstyle to dark_background (for reference, check here). Then we will set marketcolors for the chart:

Code

mc = mpf.make_marketcolors(up="g",down="r",inherit=True)
s = mpf.make_mpf_style(base_mpl_style="dark_background",marketcolors=mc)
mpf.plot(data, type="candle", volume=True, style=s,
         figsize=(6,3),title="Customised Dark theme plot on mplfinance",
        ylabel="Price [USD]")

For looking further stye changes using mplfinance, one might check their example notebook

We will move ahead and try hands on some interactive plotting libraries.

Plotly

Plotly is perhaps the most used interactive visualization library in python. It offers a great deal of customization. With Plotly, we will make two types of charts, OHLC and Volume plotted together in first one, separate in the next.

Code

import plotly.graph_objects as go
from plotly.subplots import make_subplots

fig = make_subplots(specs=[[{"secondary_y": True}]])

fig.add_trace(go.Candlestick(x=data.index, open=data["Open"],
                              high=data["High"], low=data["Low"],
                             close=data["Close"], name = "OHLC"), secondary_y=False)

fig.add_trace(go.Bar(x=data.index, y=data['Volume'], opacity=0.3,
                     name= "Volume"), secondary_y=True)

fig.update_layout(title="AAPL Candlestick with Plotly", width=900, height=400)
fig.update_yaxes(title="Price [USD]", secondary_y=False, showgrid=True)
fig.update_yaxes(title="Volume", secondary_y=True, showgrid=False)

fig.show()

In the next one, we hide the range slider for better viewing and separate the volume bar chart:

Code

fig = make_subplots(rows=2, cols=1, shared_xaxes=True, vertical_spacing=0.05, row_width=[0.2, 0.7])

fig.add_trace(go.Candlestick(x=data.index, open=data["Open"],
                              high=data["High"], low=data["Low"],
                             close=data["Close"], name = "OHLC"), row=1, col=1)

fig.add_trace(go.Bar(x=data.index, y=data['Volume'], opacity=0.3,
                     name= "Volume"), row=2, col=1)

fig.update_layout(title="AAPL Candlestick with Plotly", width=900, height=500)

fig.update(layout_xaxis_rangeslider_visible=False)
fig.show()

Plotly as well offers multiple themes and customizations, for now will use the dark theme and just change the colors of volume bars for the first plot.

Code

fig = make_subplots(specs=[[{"secondary_y": True}]])

fig.add_trace(go.Candlestick(x=data.index, open=data["Open"],
                              high=data["High"], low=data["Low"],
                             close=data["Close"], name = "OHLC"), secondary_y=False)

fig.add_trace(go.Bar(x=data.index, y=data['Volume'], opacity=0.3,
                     name= "Volume", marker_color="gray"), secondary_y=True)

fig.update_layout(title="AAPL Candlestick with Plotly", width=900, height=500, xaxis={"rangeslider":{"visible":False}}, template="plotly_dark")
fig.update_yaxes(title="Price [USD]", secondary_y=False, showgrid=True)
fig.update_yaxes(title="Volume", secondary_y=True, showgrid=False)

fig.show()

Bokeh plot

Quoting:

Bokeh is an interactive visualization library for modern web browsers. It provides elegant, concise construction of versatile graphics, and affords high-performance interactivity over large or streaming datasets. Bokeh can help anyone who would like to quickly and easily make interactive plots, dashboards, and data applications.

Bokeh additionally has great documentation and user guide. Now let’s look at a candlestick chart created using Bokeh:

Code

from bokeh.layouts import column
from bokeh.plotting import figure, show, output_notebook
from bokeh.models import ColumnDataSource, RangeTool, BoxAnnotation, HoverTool
from bokeh.resources import INLINE


output_notebook(resources=INLINE)

inc= data.Close > data.Open
dec = data.Open > data.Close
TOOLS = "pan,wheel_zoom,box_zoom,reset,save"
width_set = pd.Timedelta("16H")


fig = figure(x_axis_type="datetime", tools=TOOLS, width=900, height=350,
             title="AAPL Candlestick with Bokeh")

fig.segment(data.index, data.High, data.index, data.Low, color="black")
fig.vbar(data.index[dec], width_set, data.Open[dec], data.Close[dec], color="red")
fig.vbar(data.index[inc], width_set, data.Open[inc], data.Close[inc], color="green")
fig.yaxis.axis_label="Price [USD]"

vol = figure(x_axis_type="datetime", tools=TOOLS, width=900, height=150, x_range=fig.x_range)
vol.vbar(data.index, width=width_set, top=data.Volume, fill_color="grey", alpha=0.5)
vol.yaxis.axis_label="Volume"

show(column(fig,vol))

Loading BokehJS ...

Great! We will now change to dark theme. Check out : bokeh themes.

Code

from bokeh.plotting import curdoc
from bokeh.themes import Theme
output_notebook(resources=INLINE)

inc= data.Close > data.Open
dec = data.Open > data.Close
TOOLS = "pan,wheel_zoom,box_zoom,reset,save"
width_set = pd.Timedelta("16H")


fig = figure(x_axis_type="datetime", tools=TOOLS, width=900, height=400,
             title="AAPL Candlestick with Bokeh")

fig.segment(data.index, data.High, data.index, data.Low, color="white")
fig.vbar(data.index[dec], width_set, data.Open[dec], data.Close[dec], color="red")
fig.vbar(data.index[inc], width_set, data.Open[inc], data.Close[inc], color="green")
fig.yaxis.axis_label="Price [USD]"

doc = curdoc()
doc.theme = "dark_minimal"
doc.add_root(fig)
show(fig)

Loading BokehJS ...

That really looks good. If you are going through the code here, you can see how simple it would be to change the colors of bars. One can further add tooltips that will display the OHLC data on hovering over the candles.

One might further look at libraries like Altair, ggplot. One great library that I just found while writing this post is Holoviews

Have fun plotting!